DMRT1 Prevents Female Reprogramming in the Postnatal Mammalian Testis

LETTER doi:10.1038/nature10239

DMRT1 prevents female reprogramming in the postnatal mammalian testis

Clinton K. Matson1,2, Mark W. Murphy1, Aaron L. Sarver3, Michael D. Griswold4, Vivian J. Bardwell1,2,3 & David Zarkower1,2,3

Sex in mammals is determined in the fetal gonad by the presence or expression in Sertoli cells prevents FOXL2 expression and suggest that absence of the Y chromosome gene Sry, which controls whether Dmrt1 mutant testes become feminized during the first postnatal month. bipotential precursor cells differentiate into testicular Sertoli cells Next we examined the timing of FOXL2 induction. At postnatal day or ovarian granulosa cells1. This pivotal decision in a single gonadal (P)7, SCDmrt1KO testes had seminiferous tubules in which all Sertoli cell type ultimately controls sexual differentiation throughout the cells expressed SOX9 normally (Supplementary Fig. 2m–r), but at P14 body. Sex determination can be viewed as a battle for primacy in the fetal gonad between a male regulatory gene network in which Sry activates Sox9 and a female network involving WNT/b-catenin sig- Control ovary Dmrt1–/– (XY) Control testis nalling2. In females the primary sex-determining decision is not a bc final: loss of the FOXL2 transcription factor in adult granulosa cells can reprogram granulosa cells into Sertoli cells2. Here we show that sexual fate is also surprisingly labile in the testis: loss of the DMRT1 3 transcription factor in mouse Sertoli cells, even in adults, activates FOXL2 + DAPI Foxl2 and reprograms Sertoli cells into granulosa cells. In this Ad. P28 P28 environment, theca cells form, oestrogen is produced and germ cells SCDmrt1KO(Dhh) SCDmrt1KO(Sf1) GCDmrt1KO(Nanos3) appear feminized. Thus Dmrt1 is essential to maintain mammalian def testis determination, and competing regulatory networks maintain gonadal sex long after the fetal choice between male and female. Dmrt1 and Foxl2 are conserved throughout vertebrates4,5 and

Dmrt1-related sexual regulators are conserved throughout metazo- FOXL2 + DAPI ans3. Antagonism between Dmrt1 and Foxl2 for control of gonadal P28 P28 P28 sex may therefore extend beyond mammals. Reprogramming due to Control testis SCDmrt1KO SCDmrt1KO loss of Dmrt1 also may help explain the aetiology of human syn- gj m dromes linked to DMRT1, including disorders of sexual differenti- ation6 and testicular cancer7.

Human chromosome 9p deletions removing DMRT1 are associated SOX9 with XY male-to-female sex reversal, and Dmrt1 homologues determine sex in several non-mammalian vertebrates8–10. In mice, Dmrt1 is P14 P14 P28 expressed and required in both germ cells and Sertoli cells of the hkn testis11–13.XYDmrt1-null mutant mice are born as males with testes, although these gonads later undergo abnormal differentiation14; hence

the role of Dmrt1 in mammalian sex determination has been unclear FOXL2 (for overview of mammalian sex determination see Supplementary Fig. 1). Here we examine Dmrt1 mutant testes during postnatal P14 P14 P28 development, asking whether loss of Dmrt1 causes postnatal feminiza- ilo tion in mice. We first examined gonads of Dmrt1-null mutant males (Dmrt12/2) for the presence of FOXL2, a female-specific transcription factor expressed in granulosa cells and theca cells15,16, the two somatic cell types of the ovarian follicle (Fig. 1a). Four weeks after birth, abundant SOX9 + FOXL2 DAPI P14 P14 P28 FOXL2-positive cells were present within mutant seminiferous tubules Figure 1 | DMRT1 maintains SOX9 and suppresses FOXL2 expression in (Fig. 1b), which in control testes contain only germ cells and Sertoli postnatal Sertoli cells. a–c, FOXL2 expression detected by immunofluorescence cells (Fig. 1c). To establish the origin of the FOXL2-positive cells, we in adult (Ad.) granulosa and theca cells of control ovary (a) and intratubular cells deleted Dmrt1 either in germ cells (using Nanos3-cre) or in Sertoli cells ofDmrt1-null testis at P28 (b), but not incontrol testis (c). DAPI, 49,6-diamidino- (using Dhh-cre or Sf1-cre) (Supplementary Fig. 2a–l and Supplemen- 2-phenylindole. d–f, FOXL2 is robustly expressed when Dmrt1 is mutated in fetal Sertoli cells with Dhh-cre (d)orSf1-cre (e) but not when Dmrt1 is mutated in fetal tary Table 1). Loss of Dmrt1 in fetal Sertoli cells (SCDmrt1KO) but not germ cells with Nanos3-cre (f). g–o, Timing of FOXL2 expression. FOXL2 is in fetal germ cells (GCDmrt1KO) induced FOXL2 expression (Fig. 1d–f). absent from control testis at P14 (g–i). Cells expressing FOXL2 or FOXL2 and SCDmrt1KO gonads retained small numbers of germ cells, which SOX9 (arrowheads) are present in SCDmrt1KO testis at P14 (j–l). FOXL2- appeared to arrest in meiotic prophase on the basis of SYCP3 localiza- positive cells are abundant in SCDmrt1KO testis at P28 and most cells no longer tion (Supplementary Fig. 3). These results demonstrate that DMRT1 express SOX9 (m–o). Scale bars, 20 mm.

1Developmental Biology Center and Department of Genetics, Cell Biology, and Development, University of Minnesota, Minneapolis 55455, Minnesota, USA. 2Molecular, Cellular, Developmental Biology, and Genetics Graduate Program, University of Minnesota, Minneapolis 55455, Minnesota, USA. 3University of Minnesota Masonic Cancer Center, Minneapolis, Minnesota 55455, USA. 4School of Molecular Biosciences, Washington State University, Pullman, Washington 99164, USA.

Control testis Mutant gonad (TX) Mutant gonad (TX) Control ovary aceg DAPI

bdf h FOXL2 + SOX9 DAPI Figure 2 | Sertoli-to-granulosa transdifferentiation in the adult testis. (f, inset). g, h, FOXL2-positive cells in control ovary have DAPI morphology a–h, Expression of FOXL2 and SOX9 one month after tamoxifen (TX) injection similar to FOXL2 single-positive cells of mutant testis. FOXL2-positive cells in into Dmrt1flox/flox adult males (8 weeks and older) carrying inducible ubiquitous mutant testis resemble granulosa cells: they lack the tripartite nucleoli of Sertoli cre transgene UBC-cre/ERT2. a, b, Sertoli cells in control testis express SOX9 cells, have smaller and more rounded nuclei, and have more punctate DAPI but not FOXL2. c–f, Mutant testis has Sertoli-like cells expressing SOX9 or staining. UBC-cre/ERT2 also deletes Dmrt1 in germ cells, causing precocious SOX9 and FOXL2 (d, inset) and granulosa-like cells expressing only FOXL2 meiosis12; after one month germ cells are nearly absent. Scale bars, 20 mm. some intratubular cells co-expressed SOX9 and FOXL2 or lacked cells with typical granulosa cell nuclear morphology that lacked SOX9 SOX9 and strongly expressed FOXL2 (Fig. 1g–l). By P28 few SOX9- and strongly expressed FOXL2 (Fig. 2e–h). Thus antagonism between positive cells remained and most intratubular cells strongly expressed DMRT1 and FOXL2 continues into adulthood and Sertoli cell fate FOXL2 (Fig. 1m–o). Histological analysis of mutant gonads is shown remains plastic even after terminal differentiation. in Supplementary Fig. 4. These results show that fetal loss of Dmrt1 To evaluate further the transformation of mutant gonads, we com- causes postnatal Sertoli cells to lose the male-promoting SOX9 and pared the messenger RNA profile of control and mutant P28 testes; instead express the female-promoting FOXL2. 5,030 mRNAs were expressed .8-fold differently across this data set Loss of Foxl2 in the adult ovary can lead to transdifferentiation of or a data set comparing testis to 21 other tissues including ovary (Sup- granulosa cells to Sertoli cells2, so we asked whether loss of Dmrt1 in plementary Fig. 5a). We calculated Pearson correlation coefficients for the adult testis activates Foxl2 and causes the reciprocal sex trans- expression of these 5,030 mRNAs in mutant gonads relative to each formation, from Sertoli to granulosa. Indeed, one month after deletion tissue and found that the mutant gonad most closely resembled ovary of Dmrt1 in adult males (using a tamoxifen-inducible cre transgene), (Supplementary Fig. 5b; average R 5 0.75). Many mRNAs with we observed cells with typical Sertoli cell features including tripartite decreased expression in mutant gonads also were low in other tissues, nucleoli but expressing both SOX9 and FOXL2 (Fig. 2a–d), as well as probably reflecting a lack of male germ cells, which comprise much of the testis mass. Also, some mRNAs that were increased in mutant a gonads were increased in other tissues. Therefore, to specifically evaluate ovary-enriched mRNAs, we used bioGPS (http://www.biogps. gnf.org; see Supplementary Information) to identify 65 mRNAs with ControlSCKO(Dhh) testisSCKO(Sf1)TestisOvaryAdrenalAdiposeES cellsFetusPituitarySem.Thymus vesicleLungPlacentaSpleenMuscleKidneyDiaphramEye Saliv.Heart glandBrainSm. B.intest. marrowLiver expression closely correlated to Foxl2 and then compared their Somatic Oocyte expression in ovary relative to the other 21 tissues (Fig. 3a and Sup- 2 plementary Fig. 6). This comparison confirmed that these mRNAs are 1 highly ovary enriched. About 40% were increased in mutant gonads 0 relative to control testes; about 80% of the remainder were oocyte –1 –2 enriched. Thus loss of Dmrt1 causes large changes in mRNA expres-

Ovarian mRNAs sion, including induction of multiple ovary-enriched mRNAs. mRNA profiling of Dmrt1 mutant gonads perinatally and at P9 did not reveal Control ovary Mutant gonad Control testis bc d Figure 3 | Feminization of SCDmrt1KO XY gonads. a, Expression of ovary- enriched mRNAs with expression profiles similar to Foxl2 (see Supplementary Information). mRNAs labelled ‘somatic’ were enriched in ovarian somatic cells;

Aromatase those labelled ‘oocyte’ were enriched in female germ cells. See Supplementary Fig. 6 for higher resolution image. B. marrow, bone marrow; Saliv. gland, salivary gland; Sem. vesicle, seminal vesicle; Sm. intestine, small intestine. ef g b–d, Immunohistochemistry detection of CYP19A1/aromatase expression in follicles of control adult ovary (a) and in adult XY SCDmrt1KO gonad (b), but only in interstitial Leydig cells of control testis (c). Arrows indicate aromatase- positive granulosa cells in ovary and mutant gonad and negative Sertoli cell in control testis. Scale bars, 50 mm. e–g, Immunofluorescence detection of SMA and FOXL2. Ovarian theca cells (e, inset) are elongated cells expressing both SMA + FOXL2 DAPI hi j proteins; similar cells are present in mutant gonads (f); peritubular myoid cells in control testes express SMA and not FOXL2 (g). Scale bars, 20 mm. h–j, Immunofluorescence detection of cells coexpressing FOXL2 in the nucleus and steroidogenic enzyme CYP11A1/SCC at high levels in the cytoplasm in control ovary (h) and XY SCDmrt1KO gonads (i). SCC-positive cells in control testis (j) are interstitial Leydig cells. Mutant gonads were SCDmrt1KO(Dhh).

SCC + FOXL2 DAPI All tissues were from adult mice. Scale bars, 20 mm.

17,18 apparent feminization , consistent with the observation that FOXL2 a P28 qRT–PCR 0.003 expression starts at ,P14. 0.02 50 Further analysis of the mRNA profiling data identified highly 0.0001 Control testes (n = 2) increased expression (.5-fold, P , 0.001) of many mRNAs expressed 30 Mutant gonads (n =4 ) 0.007 0.06 in granulosa cells and required for ovarian development or function. 10 These included Foxl2, Nr5a2 (also known as Lrh1), Wnt4, LH receptor 0.02

Fold change 0.003 0.001 0.0001 (Lhcgr), prolactin receptor (Prlr), FSH receptor (Fshr), follistatin (Fst), 1 Sfrp4, Igfbp5, Inhbb, Inha and Lnfg (Supplementary Table 2). Foxl2os,a 0 noncoding RNA transcribed from the opposite strand of the Foxl2 Dmrt1 Ptgdr Sox8 Sox9 Foxl2 Esr1 Esr2 Wnt4 Rspo1 coding region, also was highly overexpressed and has been suggested b 15 P28 qChIP 0.002 as a positive regulator of Foxl2 (ref. 19). We confirmed increased 0.001 0.001 0.05 1.23×10–5 expression in mutant gonads of LRH1, a transcription factor expressed 10 20 0.0001 only in granulosa cells within the ovary and absent from the testis 0.02 (Supplementary Fig. 7a–f). Nr5a2 is probably a direct target of DMRT1 5 regulation, based on binding of DMRT1 to its promoter proximal Enrichment 0.02 sequences in vivo (Supplementary Fig. 7g). On the basis of mRNA 0 and protein expression data and changes in cellular morphology, we Ptgdr Sox8 Sox9 Foxl2 Esr1 Esr2 Wnt4 Rspo1 B2m conclude that loss of Dmrt1 in testes reprograms Sertoli cells into c Dmrt1 granulosa cells. Granulosa cells produce oestrogens, which are essential for ovarian Ptgdr development in many vertebrates; in mammals, oestrogen signalling Fgf9 also acts with FOXL2 to repress Sox9 transcription in adult granulosa 2 cells . HSD17b1 and CYP19A1/aromatase are enzymes critical for Sox9 Foxl2/Esr1,2 Wnt4 Rspo1 oestrogen synthesis, and mRNAs for both enzymes were increased Sox8 in mutant gonads (Supplementary Fig. 8). Aromatase protein is E2 B-cat robustly expressed in granulosa cells and was strongly expressed in Cyp19a1 mutant gonads (Fig. 3b–d). Consistent with these enzyme changes, oestradiol was raised in the serum of adult mutants relative to control Male-specific mRNAs Female-specific mRNAs adult males (Supplementary Information). Although expression of the Figure 4 | DMRT1 regulation of postnatal gene expression. a, qRT–PCR androgenic enzyme Hsd17b3 was not affected in mutant gonads analysis of sex-determining genes at P28. Significance of expression changes is (Supplementary Fig. 8), androgen levels were reduced based on indicated (Student’s t-test). Mutant gonads were SCDmrt1KO(Sf1); severely decreased seminal vesicle weight, a sensitive indicator of SCDmrt1KO(Dhh) mutant gonads and equivalent expression changes. androgen activity (350 6 52 mg versus 182 6 36 mg; n 5 3, P 5 0.01). b, qChIP analysis of DMRT1 DNA binding in P28 testes. Significance of Theca cells are induced during follicle growth in the ovary, probably enrichment relative to B2m (Student’s t-test) is shown. c, Model for regulation in response to granulosa cell signals21, and together with granulosa cells by postnatal sex maintenance by DMRT1. Proposed direct regulation based on and oocytes they comprise the functional unit of the ovary. Because ChIP and mRNA expression data are indicated by solid lines; indirect or mutant gonads contained apparently functional granulosa cells, we potential regulation is indicated by dashed lines. Model adapted from ref. 2. asked whether theca cells also formed. Theca cells have spindle-shaped nuclei and express both FOXL2 and smooth muscle actin (SMA) (qChIP), guided by genome-wide ChIP data from P9 testes (ChIP- 18 (Fig. 3e). Adult mutant gonads contained cells closely resembling theca chip and ChIP-seq (unpublished data)). DMRT1 bound both cells and expressing both proteins (Fig. 3f). The theca-like cells probably upstream and downstream of Sox9 and upstream of Sox8, and bound derive either from granulosa cells or peritubular myoid cells (which also weakly near Ptgdr. DMRT1 bound strongly near Foxl2, Esr1, Esr2, are elongated and express SMA; Fig. 3g). However, as seminiferous Wnt4 and Rspo1 (Fig. 4b). All of the DMRT1-associated regions tubule integrity was lost before formation of these cells (Fig. 3f and contained at least one close match to the DMRT1 DNA-binding 18,22 Supplementary Fig. 9), they could potentially derive from interstitial consensus . cells that invaded the tubule remnants. We also observed intratubular On the basis of mRNA and protein expression data and ChIP ana- cells strongly expressing the steroidogenic enzyme SCC (Fig. 3h–j); lysis, we propose a model for postnatal sex maintenance (Fig. 4b) in these cells resembled luteinized granulosa cells of the ovary (Fig. 3h), which DMRT1 maintains male fates by repressing multiple female- suggesting that granulosa cells in the mutant gonad are responsive to promoting genes and activating male-promoting genes. Sox9 is dis- gonadotropins. We therefore tested the effect of exogenous gonado- pensable for testis differentiation after sex determination24,25, suggest- tropin stimulation; treated mutants, but not controls, had additional ing that other critical male regulators remain to be found; Sox8 is a clear luteinized granulosa cells and germ cells with oocyte-like nuclear mor- candidate based on its redundancy with Sox9 (refs 23, 24). We find that phology that expressed the oocyte-specific proteins MATER and ZP2 DMRT1 represses Foxl2, which is known to maintain postnatal ovarian (Supplementary Fig. 10). This result indicates that both somatic cells fate. FOXL2 also represses Dmrt1 (ref. 2); thus antagonism between and germ cells are feminized in mutant gonads. these sex-specific transcriptional regulators may be central to sex The preceding results indicate that DMRT1 is essential for postnatal maintenance in both sexes throughout reproductive life. Wnt4 and sex maintenance. DMRT1 is a sequence-specific transcriptional regu- Rspo1 also are prime candidates for postnatal sex maintenance based lator, capable of activating or repressing transcription of target on their requirement in ovarian determination in the fetus26,27. Indeed, genes18,22. To help find targets of DMRT1 regulation with potential P28 mutant gonads had increased nuclear b-catenin in somatic cells, as roles in sex maintenance we examined expression of known fetal sex- in ovaries, but control testes did not, indicating active WNT/b-catenin determining genes in mutant gonads at P28 by quantitative polymerase signalling in the mutant gonads (Supplementary Fig. 11). Functional chain reaction with reverse transcription (qRT–PCR; Fig. 4a). Among analysis of Wnt4, Rspo1 and other known fetal sex regulators will be masculinizing genes, Ptgdr, Sox9 and Sox8, which acts redundantly with important to establish their roles in sex maintenance. Sox9 (refs 23, 24), were reduced. Among feminizing genes, Foxl2, Esr1, The analysis presented here demonstrates that deletion of Dmrt1 Esr2, Wnt4 and Rspo1 were increased. We assayed binding of DMRT1 during fetal development induces postnatal feminization of the testis, to DNA of P28 testes by quantitative chromatin immunoprecipitation causing male-to-female primary sex reversal. Moreover, deletion of

Dmrt1 in adults can reprogram differentiated Sertoli cells into appar- 8. Yoshimoto, S. et al. A W-linked DM-domain gene, DM-W, participates in primary ovary development in Xenopus laevis. Proc. Natl Acad. Sci. USA 105, 2469–2474 ent granulosa cells. Why Dmrt1 mutants are feminized only after birth (2008). remains unclear. Another male-promoting gene may act redundantly 9. Smith, C. A. et al. The avian Z-linked gene DMRT1 is required for male sex with Dmrt1 before P14, masking its function; alternatively, the testis determination in the chicken. Nature 461, 267–271 (2009). 10. Matsuda, M. et al. DMY is a Y-specific DM-domain gene required for male may lack potential feminizing activity from genes such as Foxl2 before development in the medaka fish. Nature 417, 559–563 (2002). P14. Another puzzle is that Dmrt1 mutant mice are born male, whereas 11. Kim, S., Bardwell, V. J. & Zarkower, D. Cell type-autonomous and non-autonomous human 9p deletions removing DMRT1 can cause XY feminization at requirements for Dmrt1 in postnatal testis differentiation. Dev. Biol. 307, 314–327 (2007). birth. The human sex reversal may reflect failure to maintain male sex 12. Matson, C. K. et al. The mammalian doublesex homolog DMRT1 is a transcriptional determination, and the longer human gestation may permit testis-to- gatekeeper that controls the mitosis versus meiosis decision in male germ cells. ovary reprogramming before birth. Alternatively, human testes may Dev. Cell 19, 612–624 (2010). 13. Krentz, A. D. et al. The DM domain protein DMRT1 is a dose-sensitive regulator of have potential feminizing activity earlier or may lack masculinizing fetal germ cell proliferation and pluripotency. Proc. Natl Acad. Sci. USA 106, genes redundant with DMRT1. Our results may provide insights into 22323–22328 (2009). the aetiology of human gonadal disorders, including gonadoblastoma 14. Raymond, C. S., Murphy, M. W., O’Sullivan, M. G., Bardwell, V. J. & Zarkower, D. Dmrt1, a gene related to worm and fly sexual regulators, is required for and granulosa cell tumours of the testis. Moreover, because many mammalian testis differentiation. Genes Dev. 14, 2587–2595 (2000). genes implicated in this study are evolutionarily conserved, similar 15. Schmidt, D. et al. The murine winged-helix transcription factor Foxl2 is required for mechanisms may control adult sex switching in fish and may maintain granulosa cell differentiation and ovary maintenance. Development 131, 933–942 (2004). sexual fate in the adult gonads of other vertebrates or even in other 16. Uda, M. et al. Foxl2 disruption causes mouse ovarian failure by pervasive blockage phyla. of follicle development. Hum. Mol. Genet. 13, 1171–1181 (2004). 17. Fahrioglu, U., Murphy, M. W., Zarkower, D. & Bardwell, V. J. mRNA expression METHODS SUMMARY analysis and the molecular basis of neonatal testis defects in Dmrt1 mutant mice. 12 Sex. Dev. 1, 42–58 (2007). Mouse breeding. Dmrt1 mutant and control males were generated as described ; 18. Murphy, M. W. et al. Genome-wide analysis of DNA binding and transcriptional tissue-specific Cre recombinase strains are in Supplementary Table 1. Adult wild- regulation by the mammalian Doublesex homolog DMRT1 in the juvenile testis. type or Dmrt1flox/flox females were used as controls. Mice were mixed C57BL/6J, Proc. Natl Acad. Sci. USA 107, 13360–13365 (2010). 129S1 and FVB genetic background. Protocols were approved by the Institutional 19. Cocquet, J., Pannetier, M., Fellous, M. & Veitia, R. A. Sense and antisense Foxl2 Animal Care and Use Committee. transcripts in mouse. Genomics 85, 531–541 (2005). 20. Duggavathi, R. et al. Liver receptor homolog 1 is essential for ovulation. Genes Dev. Immunofluorescence and immunohistochemistry. Immunofluorescence and 22, 1871–1876 (2008). immunohistochemistry were performed as described12. Antibodies are listed in 21. Orisaka, M., Tajima, K., Tsang, B. K. & Kotsuji, F. Oocyte-granulosa-theca cell Supplementary Table 3. Analyses included at least two biological replicates. interactions during preantral follicular development. J. Ovarian Res. 2, 9 (2009). Tamoxifen treatment. Tamoxifen-inducible deletion of Dmrt1 in adult males was 22. Murphy, M. W., Zarkower, D. & Bardwell, V. J. Vertebrate DM domain proteins bind 12 similar DNA sequences and can heterodimerize on DNA. BMC Mol. Biol. 8, 58 as described . Testes were harvested one to two months after treatment. (2007). mRNA expression analysis. mRNA expression profiling and data analysis were as 23. Chaboissier, M. C. et al. Functional analysis of Sox8 and Sox9 during sex described13 except total testis RNA was isolated from 4-week-old mice using determination in the mouse. Development 131, 1891–1901 (2004). TRIzol reagent (Invitrogen no. 15596-026). Additional detail is in Supplemen- 24. Barrionuevo, F. et al. Testis cord differentiation after the sex determination stage is tary Methods. independent of Sox9 but fails in the combined absence of Sox9 and Sox8. Dev. Biol. 12 327, 301–312 (2009). qRT–PCR. qRT–PCR was as described . qRT–PCR primers are listed in 25. Chang, H. et al. Wt1 negatively regulates b-catenin signaling during testis Supplementary Table 4. development. Development 135, 1875–1885 (2008). ChIP. ChIP followed by either microarray (ChIP-chip) or qPCR analysis (qChIP) 26. Vainio, S., Heikkila, M., Kispert, A., Chin, N. & McMahon, A. P. Female development were as described18. Gene-specific primers used for qChIP are in Supplementary in mammals is regulated by Wnt-4 signalling. Nature 397, 405–409 (1999). Table 4. 27. Parma, P. et al. R-spondin1 is essential in sex determination, skin differentiation and malignancy. Nature Genet. 38, 1304–1309 (2006). Full Methods and any associated references are available in the online version of Supplementary Information is linked to the online version of the paper at the paper at www.nature.com/nature. www.nature.com/nature. Acknowledgements We thank M. Treier for helpful discussion, K. Hatzi, A. Minkina, Received 11 February; accepted 27 May 2011. A. Peterson, the University of Minnesota Mouse Genetics Laboratory and C. Small for Published online 20 July 2011. technical assistance, J. Dean, R. Veitia and K.-i. Morohashi for antibodies, D. Greenstein and A. M. Weber-Main for comments on the manuscript, C. Manivel for histology 1. Koopman, P., Gubbay, J., Vivian, N., Goodfellow, P. & Lovell-Badge, R. Male expertise, M. Steffes and D. Gabrielson for oestradiol analysis, and the University of development of chromosomally female mice transgenic for Sry. Nature 351, Minnesota Supercomputing Institute for computational resources. This work was 117–121 (1991). funded by the NIH (GM59152), the Minnesota Medical Foundation, and a predoctoral 2. Uhlenhaut, N. H. et al. Somatic sex reprogramming of adult ovaries to testes by fellowship from the NSF (to C.K.M.). FOXL2 ablation. Cell 139, 1130–1142 (2009). 3. Raymond, C. S. et al. Evidence for evolutionary conservation of sex-determining Author Contributions C.K.M. performed mouse breeding and analysis of protein and genes. Nature 391, 691–695 (1998). mRNA expression; M.W.M. performed ChIP analysis; A.L.S. performed bioinformatic analysis; C.K.M., D.Z. and V.J.B. designed the study, analysed data, and wrote the paper; 4. Loffler, K. A., Zarkower, D. & Koopman, P. Etiology of ovarian failure in M.D.G. provided mRNA profiling expertise; all authors discussed the results and edited blepharophimosis ptosis epicanthus inversus syndrome: FOXL2 is a conserved, the paper. early-acting gene in vertebrate ovarian development. Endocrinology 144, 3237–3243 (2003). Author Information mRNA expression profiling data have been deposited at the Gene 5. Raymond, C. S., Kettlewell, J. R., Hirsch, B., Bardwell, V. J. & Zarkower, D. Expression Expression Omnibus under accession number GSE27261 and can be reviewed at of Dmrt1 in the genital ridge of mouse and chicken embryos suggests a role in http://www.ncbi.nlm.nih.gov/geo/query/ vertebrate sexual development. Dev. Biol. 215, 208–220 (1999). acc.cgi?token5zlctbiugsamymtc&acc5GSE27261. Reprints and permissions 6. Tannour-Louet, M. et al. Identification of de novo copy number variants associated information is available at www.nature.com/reprints. The authors declare no with human disorders of sexual development. PLoS ONE 5, e15392 (2010). competing financial interests. Readers are welcome to comment on the online version 7. Turnbull, C. et al. Variants near DMRT1, TERT and ATF7IP are associated with of this article at www.nature.com/nature. Correspondence and requests for materials testicular germ cell cancer. Nature Genet. 42, 604–607 (2010). should be addressed to D.Z. ([email protected]) or V.J.B. ([email protected]).

METHODS the data; (2) filter the data set for genes that showed at least three observations with Mouse breeding. Conditional Dmrt1 mutant and control males were generated as abs(val) . 5 3 (eightfold), which resulted in 5,030 genes passing the filter using described12; tissue-specific Cre recombinase strains are in Supplementary Table 1. both data sets combined; and (3) cluster the data on the gene axis using average Adult wild-type or Dmrt1flox/flox females were used as controls. Mice were of mixed linkage hierarchial clustering. The experimental axis was defined by order of decreasing correlation to the mutant testes calculated as described earlier. C57BL/6J, 129S1 and FVB genetic background. Protocols were approved by the 31 University of Minnesota Institutional Animal Care and Use Committee. Javatreeview Software was used to generate heatmap images. 12 Immunofluorescence and immunohistochemistry. Both immunofluorescence qRT–PCR. qRT–PCR was performed as described . qRT–PCR primers are listed and immunohistochemistry were performed as described12.Antibodiesarelistedin in Supplementary Table 4. Supplementary Table 3. Analyses included a minimum of two biological replicates. Chromatin immunoprecipitation. ChIP followed by either microarray (ChIP- 18 Tamoxifen treatment. Tamoxifen-inducible deletion of Dmrt1 in adult males was chip) or qPCR analysis (qChIP) were performed as described . Gene-specific performed as previously described12. Testes were harvested one to two months primers used for qChIP are in Supplementary Table 4. after treatment. Oestadiol assays. Serum oestradiol was assayed using a clinical electrochemilu- mRNA expression analysis. mRNA expression profiling and data analysis were minescence immunoassay (Roche Estradiol II, 03000079 122) according to man- performed as described13 except total testis RNA was isolated from 4-week-old ufacturer’s instructions. Three of three males assayed had levels below the detection limit, whereas two of three females had measurable oestradiol (5.0 mice using TRIzol reagent (Invitrogen no. 15596-026). Affymetrix Mouse Genome 21 28 and 19.7 pg dl ). Two of three SCDmrt1KO(Dhh) mutant males had measurable 439 2.0 arrays were normalized by GC-RMA normalization using GeneData 21 Refiner. The Raw .cel files and the normalized data are deposited in the Gene oestradiol (5.6 and 21.2 pg dl ). Expression Omnibus (GEO)29 under accession number GSE27261. GSE9954 Gonadotropin treatment. Six-to-eight-week-old mutant males, control males was obtained from the GEO database. The arrays with the highest sample iden- and control females were treated with 5 units of pregnant mare serum by intra- tification numbers were removed from the tissue data set to select 22 tissue types, peritoneal injection and gonads were harvested 48 h later. each with three experimental replicates. When multiple probe sets were mapped to 28. Wu, J., Irazarry, R. A., Gentleman, R., Martinez-Murillo, F. & Spencer, F. A model- the same gene symbol, these values were averaged to obtain one value for each gene based background adjustment for oligonucleotide expression arrays. J. Am. Stat. symbol. Direct Pearson correlation R values were calculated using all array data Assoc. 99, 909–917 (2004). following reduction to gene symbols, and these values are shown in Fig. 2b. 29. Edgar, R., Domrachev, M. & Lash, A. E. Gene Expression Omnibus: NCBI gene Each experiment in our data set was divided by the average expression value expression and hybridization array data repository. Nucleic Acids Res. 30, from control testis tissue. GSE9954 data were separately divided by the average 207–210 (2002). 30. de Hoon, M. J., Imoto, S., Nolan, J. & Miyano, S. Open source clustering software. signal obtained from the GSE9954 testis samples. This was done separately for Bioinformatics 20, 1453–1454 (2004). each data set to determine how samples from each data set differed from a baseline 31. Saldanha, A. J. Java Treeview—extensible visualization of microarray data. ‘testis’ expression state. Cluster 3.0 software30 was used to: (1) log base 2 transform Bioinformatics 20, 3246–3248 (2004).

ERRATUM doi:10.1038/nature10411 DMRT1 prevents female reprogramming in the postnatal mammalian testis Clinton K. Matson, Mark W. Murphy, Aaron L. Sarver, Michael D. Griswold, Vivian J. Bardwell & David Zarkower

Nature 476, 101–104 (2011) In Fig. 3 of this Letter, the label above Fig. 3c should have read ‘‘XY Dmrt1KO gonads’’ should have read ‘‘XY SCDmrt1KO gonads’’. ‘‘Mutant gonad’’ not ‘‘Mutant gonad (TX)’’. In the legend to Fig. 3i These errors were corrected online on 3 August 2011.